Abstract One in four deaths in the United States is caused by cardiovascular disease?an enormous public health burden. An under-appreciated risk factor for heart disease is obstructive sleep apnea (OSA). In OSA, the upper airway collapses repeatedly during sleep and prevents breathing. We have found that the duration of these respiratory events is heritable and is a risk factor for heart disease and death. We believe that the time course of breathing disturbances through the night is a rich source of information about a patient's OSA severity and the patient's long term risk for heart disease. OSA severity is currently defined by the mean number of respiratory events per hour asleep (apnea-hypopnea index, AHI). This number ignores physiologically significant variation in event duration, their spacing, and association with different sleep stages. Moreover, current clinical cutoffs for mild, moderate, and severe OSA are not based on any physiological mechanism. We have therefore analyzed two physiologically informative parameters?the duration of respiratory events and their clustering within the night?and have found that those with short and regularly occurring respiratory events are at the greatest risk of dying. Event duration is the most heritable of OSA traits, suggesting a potential genetic underpinning to this phenotype. Our goals in this project are to determine how respiratory event duration and inter-event variability predict future cardiovascular disease using prospective data sets available through the National Sleep Research Resource. We will test whether these novel OSA metrics predict risk in multiple independent cohorts to establish their generalizability, and we will determine whether these metrics help stratify differential risk between men and women. To date, the AHI has not been shown to be a good predictor of future risk in women, yet therapeutic management for women continues to be guided by this single number. Identifying better predictors for women from the information contained in the night-time polysomnogram has the potential to dramatically change the therapeutic strategies for women with OSA. Finally, in cross-sectional studies, we will test whether subjects with short regularly occurring respiratory events have elevated markers of cardiovascular risk, based on state-of-the-art imaging measures of cardiac function and coronary artery calcification.